The long term aim of this project is to develop a polymorphism detection system consisting of: 1) A DNA and polymorphism isolation system for large genomic fragments; 2) a random DNA array preparation kit for arraying of selected polymorphisms for parallel analysis; 3) A universal-probe hybridization and ligation system for assaying random DNA arrays; and 4) Data collection and DNA sequence assembly and analysis software. The system will have the capacity to discover in one assay polymorphisms in over 10 megabases of targeted gene sequence in thousands of individual DNA samples. The specific aims of this proposal are: 1) Develop and demonstrate a method for the selective isolation of long intact DNA for targeted genomic regions, and then use these molecules to demonstrate random DNA array preparation, and 2) Develop and demonstrate a method for DNA-circularization-based selection of short DNA segments having sequence differences between reference and test DNA preparations and demonstrate the identification of these sequence differences on random arrays. The isolation of long DNA fragments in the order of 100 kilobases in size will allow coverage of large regions of the genome with minimal reagents. Nucleotide mismatches between isolated genomic DNA and reference DNA will be used to capture small regions of sequence surrounding the polymorphisms for amplification and arraying, thus increasing efficiency of the analysis over 20 fold. High throughput parallel characterization of the polymorphisms in thousands of individuals will then proceed utilizing sequencing-by-hybridization on random arrays of isolated DNA fragments. Once fully developed, the proposed system will be applicable to the comprehensive polymorphism detection of large genomic intervals in one assay for both diagnostic and research applications including the screening of inherited genetic mutations or somatic mutations that may predispose to disease. Polymorphism discovery and screening will also be important for understanding how individuals respond to medications and which course of treatment will be most beneficial and cost effective. [unreadable] [unreadable]